Asthma is a problem of increasing importance in the United States, especially among minority and inner-city groups. Causes for this increase in asthma are poorly defined; however, studies from many areas of the United States point to a high prevalence of sensitization to cockroach among asthmatics, especially in inner cities, but also among rural and small town residents. Skin test surveys, bronchial challenge studies, and evidence of occupational illness demonstrate the relevance of cockroach allergen to asthma. Cockroach allergen may be found in vacuumed dust from households, but a critical level beyond which sensitization or symptoms are expected is not known. The amount and nature of any cockroach allergen suspended in household air is not known. The effect of suggested cockroach control measures on allergen in household dust and air is not known. Similarly, it is not known if reduction of exposure to cockroach allergen will have an effect on lung function in cockroach-sensitive asthmatics. This proposed study will investigate several of these gaps in knowledge of cockroach allergens and their relation to asthma. Hypotheses to be tested are: (1). Simple control of cockroach population in a detached dwelling will lead to a decrease in cockroach allergen in vacuumed dust. (2). This reduction in cockroach allergen will have a detectable beneficial effect on the status of asthma in patients living in the home. A parallel group, placebo-controlled trial of the use of hydramethylnon bait stations will be used to confirm, firstly, that this intervention reduces cockroach population (as expected from the literature), and, secondly, that cockroach allergen in vacuumed dust as measured by specific monoclonal antibodies also decreases. The effect on asthma will subsequently be determined by treating, also in a parallel group, double- blinded, placebo-controlled fashion, the homes of patients with asthma and cockroach sensitivity, attempting to control for the important confounding variables, especially other indoor allergens. Sample size calculation suggests that, with methacholine reactivity as the outcome indicator, sufficient power can be achieved with a realistic sample size to detect a clinically important improvement in asthma. Proof of the efficacy of this inexpensive intervention in reducing asthma severity could have major public health implications.